Re: [Koha] Importing data into MARC
Tuesday, August 31, 2004 17:48 CDT Hi, Andres, First off, kudos on what you were saying yesterday regarding salaries. I couldn't have agreed with you more. As for your databases, I suppose the solution you would want would bring the copy specific information in the second database into the title specific records of the first. Perhaps someone else on the listserv can suggest/help write a script that could merge the databases. How many records are we talking about? And how many fields of information do you want/need preserved from the second database with the purchase info? In terms of MARC coding solutions, would the 852 field suffice alone? The $z subfield is repeating. There is also the 541 note which is repeating and could be incorporated to contain more copy-specific information with defined subfields. Check that out at URL <http://www.loc.gov/marc/bibliographic/ecbdnot2.html#mrcb541>. Take a look at the Holdings information at URL <http://www.loc.gov/marc/holdings/echditem.html> ; this gives guidance on embedding holdings information, i.e. including an 876 note within each MARC record for an item. You would be repeating the 852 field (and/or 541) for each different copy of the item within the item record and (keeping in mind that I've never had to do this but from my reading) you would be linking these to their corresponding 876 notes (which have enough fields available for fairly detailed acquisitions data) using the $8 to clarify which item was meant (I suppose the bar code would be the natural choice). I hope this helps a bit. Let us know how things progress. Cheers, Steven F. Baljkas library tech at large Koha neophyte Winnipeg, MB, Canada
From: Andres Tarallo <tarallo@ort.edu.uy> Date: 2004/08/31 Tue PM 04:59:22 CDT To: koha@lists.katipo.co.nz Subject: [Koha] Importing data into MARC
We're currently finishing our mapping to import our database (CEPAL catalog in DOS/ISIS) to Koha. We have a question, and we want to know the best solution to feed data for bulkmarcimport.pl.
Our library has two databases. The first one has a detailed description of each title avaliable in the library; this database has one record for each title we have, no matter how many copies of it we have. The second one has all the purchase information, in this database you have one record for every copy of a book you have. Both databases are releted via a pointer in the first one that references the records of the second one.
Now, we don't want to lose information and we're facing that many of our books have many copies, so we want to know whats the best way to build our MARC records, A record with full information from both databases for each book we have?
Thanks for your feedback
Andres _______________________________________________ Koha mailing list Koha@lists.katipo.co.nz http://lists.katipo.co.nz/mailman/listinfo/koha
Baljkas Family wrote:
[..]
As for your databases, I suppose the solution you would want would bring the copy specific information in the second database into the title specific records of the first.
The first idea that came to my mind was merging records from the Title databse with the records i have in the databse were i have purchase information.
Perhaps someone else on the listserv can suggest/help write a script that could merge the databases.
I'm currently working on that, i was a litle delyaed because for some reason one of our databases had some garbage that prevented me from reading correctly the record, now i'm on track.
How many records are we talking about? And how many fields of information do you want/need preserved from the second database with the purchase info?
i have to libraries with about 30.000 record each. The titles datasbes has an average of 21.000 records the purchase database has about 30000, meny titles are repeated and we just keep there five or six fields that are relevant for purchases.
In terms of MARC coding solutions, would the 852 field suffice alone? The $z subfield is repeating.
Maybe, I'm not a MARC expert, we need to keep track of the purchase information.
There is also the 541 note which is repeating and could be incorporated to contain more copy-specific information with defined subfields. Check that out at URL <http://www.loc.gov/marc/bibliographic/ecbdnot2.html#mrcb541>.
It's becoming like my pillow :), at the same time I have to map our current Isisi datasbe in CEPAL layout to Marc21.
Take a look at the Holdings information at URL <http://www.loc.gov/marc/holdings/echditem.html> ; this gives guidance on embedding holdings information, i.e. including an 876 note within each MARC record for an item.
You would be repeating the 852 field (and/or 541) for each different copy of the item within the item record and (keeping in mind that I've never had to do this but from my reading) you would be linking these to their corresponding 876 notes (which have enough fields available for fairly detailed acquisitions data) using the $8 to clarify which item was meant (I suppose the bar code would be the natural choice).
This week I'm working in mapping CEPAL ==> MARC21, with some help from the librarians. We were somewhat stuck becouse in CEPAL laoyuot we have a filed coded 002 were we have an inventory number: it seems to us that this maps to field 035 of MARC21. You'll here from me soon. Andres
participants (2)
-
Andres Tarallo -
Baljkas Family